Research map / Detection and segmentation / CNN features
Towards Accurate Multi-Person Pose Estimation in the Wild
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add in-house labeled training data | G-RMI · COCO test-dev | AP: 0.649 -> 0.685 | Helped |
| Add in-house labeled training data | G-RMI · COCO test-standard | AP: 0.643 -> 0.673 | Helped |
| Use larger 800x1200 detector input instead of 600x900 (ablation) | G-RMI with COCO-only pose training · COCO mini-val | AP: 0.657 -> 0.667 | Helped |
| Use larger 800x1200 detector input instead of 600x900 (ablation) | G-RMI with COCO+int pose training · COCO mini-val | AP: 0.693 -> 0.700 | Helped |
| Replace predicted person boxes with ground-truth boxes (ablation) | G-RMI with COCO-only pose training · COCO mini-val | AP: 0.667 -> 0.704 | Helped |
| Replace predicted person boxes with ground-truth boxes (ablation) | G-RMI with COCO+int pose training · COCO mini-val | AP: 0.700 -> 0.745 | Helped |
| Use ResNet-101 instead of ResNet-50 at 353x257 crop size (ablation) | G-RMI pose estimator · COCO test-dev | AP: 0.666 -> 0.685 | Helped |
| Increase crop size from 257x185 to 353x257 with ResNet-101 (ablation) | G-RMI pose estimator · COCO test-dev | AP: 0.661 -> 0.685 | Helped |
| Set OKS-NMS threshold to 0.3 rather than 0.5 (ablation) | G-RMI with COCO-only pose training · COCO mini-val | AP: 0.667 -> 0.664 | No clear effect |
| Vary OKS-NMS threshold from 0.3 to 0.9 (ablation) | G-RMI with COCO+int pose training · COCO mini-val | AP: 0.699, 0.700, 0.701, 0.694 | No clear effect |
| Rank pose proposals using keypoint confidence rather than Faster-RCNN box scores | G-RMI · COCO keypoint detection | AP | Helped |
| Use keypoint-based OKS-NMS rather than box-level overlap NMS for pose detections | G-RMI · COCO keypoint detection | AP | Helped |
Closest papers by meaning
- Simple Pose: Rethinking and Improving a Bottom-up Approach for Multi-Person Pose Estimation (AAAI 2020 · similarity 0.92)
- FCPose: Fully Convolutional Multi-Person Pose Estimation With Dynamic Instance-Aware Convolutions (CVPR 2021 · similarity 0.90)
- Cascaded Pyramid Network for Multi-Person Pose Estimation (CVPR 2018 · similarity 0.90)
- Explicit Box Detection Unifies End-to-End Multi-Person Pose Estimation (ICLR 2023 · similarity 0.89)
- ProbPose: A Probabilistic Approach to 2D Human Pose Estimation (CVPR 2025 · similarity 0.87)
- On the Calibration of Human Pose Estimation (ICML 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.